What wolf pups that play fetch reveal about your dog

three were responsive and not fully responsive at that.”

And based on her personal experience as a dog owner, she said: “Many dogs will never return a ball.”

Hansen Wheat’s point is not that ball fetching itself is important, but that it shows a kind of social engagement that she didn’t expect to see in wolf puppies. Ostrander acknowledges that, but still wonders whether ball fetching is the best trait to study.

Hansen Wheat and Temrin tested three litters of handraised wolf pups, the first in 2014 and two others in 2015 an 2016. They used a test that is standard in Sweden to assess a range of behaviors in dog puppies, like sociability, curiosity and playfulness. The researchers had an experienced “puppy assessor,” a stranger to the puppies, administer the test, while they observed and recorded the behavior.

In one part of the test the assessor threw a ball, waited to see if the puppy played with it and then called to the puppy to bring it back. Each pup got three tries. Until the 2016 litter only one pup even paid the ball any attention. But in the 2016 litter, three of six wolves brought the ball back to the tester. Each pup had three tries. One did it all three times. The other two fetched the ball two out of three tries.

Hansen Wheat would very much like to get more data, on more wolves, but she said that the research with just a few pups was enough to show that the innate behavior was present in some wolves and was probably present in dog ancestors.

To understand what happened in domestication, researchers try to separate out traits that resulted from new mutations as humans created new kinds of dogs, and traits that were already present to some extent in ancient wolves.

The short legs of Corgis, bulldogs and dachshunds, or the flat faces of pugs, are the result of mutations that occurred while humans were developing new dog breeds. Another dog ability, to digest starch better than wolves, developed in a more complicated way. It resulted from an increase in the number of copies of a specific gene. Even among dog ancestors, the gene seems to have been present in different numbers of copies in different animals, implying that it was not related to domestication.

Karlsson said that the study of domestication needs “more research like this new paper.” And, she added, since it has been a very short evolutionary time since dogs and wolves split, “Our assumption should be that dogs and wolves won’t differ.”